Report Reveals that Human Costs of War Have Shrunk Dramatically
20 Jan 2010
Findings stand in sharp contrast to prevalent media images of contemporary warfare
Claims of 5.4 million war deaths in the Democratic Republic of Congo rejected
NEW YORK—Challenging commonsense assumptions, a new study from the Canadian research team that produced the much-cited Human Security Report, reveals that nationwide death rates actually fall during the course of most of today’s armed conflicts.
The new study, The Shrinking Costs of War, was produced by the Human Security Report Project at the School for International Studies at Simon Fraser University and funded by the governments of the United Kingdom, Norway, Sweden, and Switzerland. It will appear in the forthcoming Human Security Report 2009 to be published by Oxford University Press.
The Shrinking Costs of War argues that wartime mortality, from disease and malnutrition, as well as war-inflicted injuries, has been driven downwards by:
- Significant changes in the nature of warfare––evident in the 70 percent decline in the number of high-intensity conflicts since the end of the Cold War.
- More than 30 years of highly effective health interventions in poor countries in peacetime––which have cut death tolls from disease during wartime.
- A dramatic increase in the level and effectiveness of humanitarian assistance to people in war zones.
These findings stand in sharp contrast to contested claims of enormous death tolls in Iraq, Darfur, and the Democratic Republic of the Congo (DRC).
The Shrinking Costs of War also provides the most comprehensive analysis to date of the claim that 5.4 million people have died because of the war in the DRC. The study argues that the true death toll is far smaller.
The Paradox of Mortality Rates that Decline in Wartime
Four years after the first Human Security Report was launched, it is increasingly accepted that deaths from wartime violence have decreased. But in poor countries where most wars are now waged, violent deaths make up only a small percentage of the total number of conflict deaths.
Most deaths in today’s conflicts are caused by war-exacerbated disease and malnutrition, not war-inflicted injuries. Long ignored, these “indirect deaths” have attracted increasing attention in recent years in the wake of media and NGO revelations of the massive death tolls from war-driven disease and hunger in Darfur, the DRC and elsewhere.
The most surprising and counterintuitive finding in The Shrinking Costs of War is that mortality rates mostly go down during war. Yet the facts are indisputable. In 2009, Human Security Report Project researchers examined under-five mortality rates in 18 war-affected countries in sub-Saharan Africa between 1970 and 2007. They found that these rates declined overall during periods of warfare in some 80 percent of the countries in question. A World Bank study published in 2008 found that globally, median adult and infant mortality rates also declined during wartime.
“No one, of course, is suggesting that war is good for people’s health,” says Human Security Report Project Director, Andrew Mack. “But the reality is that the death toll in most of today’s wars is too small to reverse the steady decline in peacetime mortality that developing countries have been experiencing for more than thirty years.”
The decline in peacetime mortality has been dramatic. Under-five mortality––for which we have the best data––has fallen worldwide by some 60 percent since 1960. Much of the improvement has been due to the widespread introduction of low-cost, but highly effective lifesaving health interventions––notably immunization against a number of deadly diseases that have killed tens of millions of children in the developing world.
The most important consequence of these changes for this study is that immunizing children in peacetime dramatically reduces their risk of succumbing to disease in wartime. As immunization coverage goes up, child mortality rates in war zones go down.
Since few wars engulf entire populations, immunization coverage can, and often does, increase during wartime. In the DRC, for example, measles immunization coverage was at some 20 percent in 1998––the year the war started. By 2007, it was almost 80 percent. The increase in diptheria, pertussis and tetanus immunization coverage was very similar.
Humanitarian assistance has also played an important role in driving down wartime mortality rates. “There has been a more than three-fold increase in the level of humanitarian assistance per displaced person since the end of the Cold War. It has become more effective and saved countless lives,” says Professor Mack.
But the most important driver of the dramatic decline in wartime mortality, according to The Shrinking Costs of War, has been the changing nature of warfare––from wars fought with huge armies, heavy conventional weapons, and the intervention of the major powers during the Cold War era, to the “low intensity insurgencies” of the past two decades fought mostly by small, lightly armed rebel groups that seek to avoid major battles. Although often noted for their savage attacks against civilians, these conflicts kill relatively few people.
An additional consequence of the changing nature of war is that fighting has become more localized. One recent study on Africa found that, on average, little more than 10 percent of the territory of countries at war is affected by serious violence. In the areas that remain relatively violence-free, health services will often still be provided and livelihoods maintained––both with positive outcomes for population health.
The cumulative effect of all of these changes has been profound. “The average war in the new millennium generates 90 percent fewer battle deaths than did the average war in the 1950s,” Mack points out. And the evidence in The Shrinking Costs of War indicates that indirect war deaths have been declining even more steeply than battle deaths.
There are still wars and instances of mass violence against civilians so deadly that they do, in fact, reverse the long-term decline in mortality rates. The Rwandan genocide is perhaps the most dramatic recent case in point. But some of the most publicized claims about huge war death tolls over the last five years––in Iraq, Darfur and, most recently, the DRC––have been sharply contested.
Challenging Death Toll Estimates in the Democratic Republic of Congo
The International Rescue Committee’s (IRC) research project to determine how many people have died since 1998 because of the war in the DRC has been the most ambitious survey-based study ever undertaken to estimate “excess” war deaths––those that would not have occurred had there been no war.
The IRC’s claims of shockingly high death tolls in the Congo focused the world’s attention on a largely forgotten war and helped build international support for increased humanitarian assistance and a greater peacekeeping presence in the violence-prone east of the country.
But there are major problems with the methodology and some of the key assumptions that underpin the IRC’s 5.4 million death toll estimate.
First, the IRC’s estimate of the child mortality rate in the DRC is double that of the well-regarded Demographic and Health Survey. Both cannot be correct. Child mortality is an important indicator of the overall war death toll in the Congo since the IRC’s data show that nearly half of all excess deaths in the country were of children under-five.
Second, in its initial two surveys the IRC failed to follow standard survey practice to ensure that the population in the areas surveyed were representative of the larger population of the war-affected eastern part of the country.
Although the survey data made it clear that mortality levels in parts of the region were very high, the fact that the sample was not representative of the region as a whole means that no confidence can be placed in the IRC’s excess mortality estimates from this period.
Third, even if this critical misstep is ignored, other methodological errors led to large and unwarranted increases of the excess death estimates. For example, when a series of erroneous assumptions in the first survey were rectified by the Human Security Report Project researchers, the estimated death toll dropped from 1.6 million to some 680,000––a 60 percent decline.
Finally, the IRC’s reliance on an inappropriately low pre-war mortality rate for all five surveys distorted its results very substantially. In the last three surveys, for example, the excess death estimate of some 2.8 million shrinks dramatically––to less than 900,000––when a more appropriate pre-war mortality rate is used.
The evidence suggests, in other words, that for the last three surveys, the only ones that covered the entire country and where the sample populations were appropriately selected, the IRC’s excess death estimate is more than three times too large.
The problems involved in estimating excess death are by no means unique to the IRC and the DRC. Indeed The Shrinking Costs of War devotes a whole chapter to arguing that attempts to estimate war deaths using mortality surveys confront methodological challenges so great that they can rarely succeed.
The Shrinking Costs of War
The international community can take little credit for the decline in war deaths that has resulted from recent changes in the nature of warfare. However, there is little doubt that the highly successful international drive to reduce mortality in the developing world in peacetime, has also played an important role in reducing death tolls in wartime––as has the greatly increased provision, scope and effectiveness of humanitarian assistance.
The international community can also claim some credit for the fact that the number of conflicts being fought around the world has declined by some 40 percent since the end of the Cold War. This fact, along with the worldwide decline in the deadliness of individual wars, leaves little doubt that the human costs of war––from disease and malnutrition as well as war-related injuries––have shrunk dramatically.
2 December 2010: A new version of the Shrinking Costs of War was released in the Human Security Report 2009/2010.
« All news